The TI E2E™ design support forums will undergo maintenance from July 11 to July 13. If you need design support during this time, open a new support request with our customer support center.

  • State TI Thinks Resolved
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 10 subscribers
  • Views 172 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

ADS58C20: ADS58C20: Question about 1.9Vpp differential input not reaching digital full scale?

benson luo
Prodigy 70 points
Part Number: ADS58C20

Tool/software:

hi，

The ADS58C20 specification page 48 describes: "For a full-scale differential input, each input pin INP, INM has to swing symmetrically between VCM + 0.45V and VCM – 0.45V, resulting in a 1.9VPP differential input swing."

From the specification, I understand that the INP and INM inputs swing between 1.9+ 0.45V and 1.9 – 0.45V, and the 1.9Vpp (0.45*4) differential input can reach the full scale of the ADC.
Why is our actual test inconsistent with the specification description:

1. Test environment:

The test environment is set up as shown in the figure below，The signal generator output impedance is 50Ω, 30MHz sine,

1.9Vpp differential signal, input to ADC (14bit) through a 0.1uF DC blocking capacitor, the collected digital signal is only 4000, which does not reach the digital full scale of 14bit (8192);


At this time, continue to increase the input signal amplitude to 3.8Vpp, and the digital signal full scale can reach 8192 (2 to the 13th power)

2. Problem points:

1) The test situation is inconsistent with the description in the specification book. The differential 1.9Vpp input does not reach the digital full scale (only half of it, about 4000);

2) The differential input is 3.8Vpp (1.9Vpp*2), and the ADC reaches the digital full scale 8192 output. Is the description in the specification book wrong, or are there other problems?

Hope to receive your reply as soon as possible!

Thank you for your support! !

  • 0
    TI__Guru* 76624 points

    Hello,

    We may have ADCs who can achieve better performance. Could we ask our marketing team to reach out to you on new products?

    The differential voltage is defined as Vss basically. 

    Please advise if the customer is using decimation by 2 mode to unlock the 14-bit mode.

    Please also check your transformer/balun bandwidth to see the 3dB point of the balun. 30Mhz is low frequency for most of the baluns, and you may need to use a different balun to support 30Mhz IF.

  • 0 in reply to Kang Hsia
    Prodigy 70 points

    Thank you for your prompt reply and suggestions

    I can confirm that the test environment is fine, and the differential signal amplitude input to the ADC is correct through the oscilloscope differential probe (800M);
    Please also answer my questions directly:
    1. For AD58C20, why can't the differential input Vpp 1.9V reach the digital full scale, which is inconsistent with the specification
    2. Can the differential signal input be set to Vpp3.8V to fully utilize the dynamic range of the full scale?

    Thank you again!

  • 0 in reply to benson luo
    TI__Guru* 76624 points

    Hello Benson,

    I believe I have already answered all of your questions. Please forgive me if I am not sure what is missing. I have included all the diagrams above. Please read carefully.

    Links said:
    The key question is whether the differential input dynamic range of ADC is 1.9Vpp or 3.8Vpp as claimed in the specification book?

    Please see below diagram and highlighted area.

    VA = 1.9V + 0.47V, VB = 1.9V -0.47V. VID = 0.47-(-0.47) = 0.92V. VSS = 2*VID = 1.94V.

    Kang Hsia said:

    The differential voltage is defined as Vss basically. 

    PLEASE answer my other debug points:

    Kang Hsia said:
    Please advise if the customer is using decimation by 2 mode to unlock the 14-bit mode.

    Kang Hsia said:
    Please also check your transformer/balun bandwidth to see the 3dB point of the balun. 30Mhz is low frequency for most of the baluns, and you may need to use a different balun to support 30Mhz IF.
  • 0 in reply to Kang Hsia
    Prodigy 70 points

    Thanks for your reply!

    It is certain that the signal amplitude after balun is normal. As mentioned above, the input amplitude of ADC has been verified to be in line with expectations by using differential probe;

    "decimation by 2 mode to unlock the 14-bit mode." is still under confirmation. What is the impact here?

    Putting aside other questions, I just want to know:

    1. Is the datasheet describing that the differential input Vpp of ADS58C20  is 1.9Vpp accurate?

    2. If it is correct, why does 1.9Vpp not reach the digital full scale（only more than 4000）？, and differential input 3.8Vpp can reach it（8192）

  • 0 in reply to benson luo
    TI__Guru* 76624 points
    benson luo said:
    "decimation by 2 mode to unlock the 14-bit mode." is still under confirmation. What is the impact here?

    Only this mode allows you to see all 14-bit. Otherwise, other modes yields 9 bit only due to export restriction. This is why I said this part may not be the best suitable part for your application.

    benson luo said:
    1. Is the datasheet describing that the differential input Vpp of ADS58C20  is 1.9Vpp accurate?

    Yes, it is correct.

    benson luo said:
    2. If it is correct, why does 1.9Vpp not reach the digital full scale（only more than 4000）？, and differential input 3.8Vpp can reach it（8192）

    This is why I asked you if you are in decimation by 2 mode.

  • 0 in reply to Kang Hsia
    Prodigy 70 points

    Thanks again for your reply

    1. We are using decimation by 2 mode, 14bit

    2. Is the differential input limited to 1.9V to ensure that the performance of indicators such as SNR, SFDR, and SINAD meet the specifications? If it exceeds 1.9V, does it mean that these indicators will deteriorate?

  • 0 in reply to benson luo
    Prodigy 70 points

    Sorry to bother you again, please answer the remaining questions.

    Thank you very much！